sedrickkeh commited on
Commit
a9e1044
1 Parent(s): 9e1ae12

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: hp_ablations_mistral_epoch5_dcftv1.2
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_mistral_epoch5_dcftv1.2
17
 
18
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.4100
21
 
 
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: hp_ablations_mistral_epoch5_dcftv1.2
 
16
 
17
  # hp_ablations_mistral_epoch5_dcftv1.2
18
 
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v1.2_no-curation_gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.4100
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_loss": 0.4100167155265808,
4
+ "eval_runtime": 96.3977,
5
+ "eval_samples_per_second": 103.374,
6
+ "eval_steps_per_second": 0.405,
7
+ "total_flos": 3097981385441280.0,
8
+ "train_loss": 0.3954814358659693,
9
+ "train_runtime": 27395.2752,
10
+ "train_samples_per_second": 34.555,
11
+ "train_steps_per_second": 0.068
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_loss": 0.4100167155265808,
4
+ "eval_runtime": 96.3977,
5
+ "eval_samples_per_second": 103.374,
6
+ "eval_steps_per_second": 0.405
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "total_flos": 3097981385441280.0,
4
+ "train_loss": 0.3954814358659693,
5
+ "train_runtime": 27395.2752,
6
+ "train_samples_per_second": 34.555,
7
+ "train_steps_per_second": 0.068
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1377 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1850,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02702702702702703,
13
+ "grad_norm": 3.48477044162232,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8622,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.05405405405405406,
20
+ "grad_norm": 2.360261766882162,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.6871,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.08108108108108109,
27
+ "grad_norm": 2.861337031562045,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.6496,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.10810810810810811,
34
+ "grad_norm": 2.110955586370757,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6379,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.13513513513513514,
41
+ "grad_norm": 2.6603965950552375,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6224,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.16216216216216217,
48
+ "grad_norm": 2.3295005659015637,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6158,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.1891891891891892,
55
+ "grad_norm": 1.4436743736286393,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6121,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.21621621621621623,
62
+ "grad_norm": 1.480185038939668,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6048,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.24324324324324326,
69
+ "grad_norm": 1.7946241900816966,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6027,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.2702702702702703,
76
+ "grad_norm": 2.953041989671829,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6057,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.2972972972972973,
83
+ "grad_norm": 2.3672617838763292,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.603,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.32432432432432434,
90
+ "grad_norm": 2.570180323617788,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.5968,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.35135135135135137,
97
+ "grad_norm": 2.4395069709384405,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.5971,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3783783783783784,
104
+ "grad_norm": 2.139806105177032,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.5967,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.40540540540540543,
111
+ "grad_norm": 1.8058881040832764,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.588,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.43243243243243246,
118
+ "grad_norm": 1.906112223883072,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.5911,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4594594594594595,
125
+ "grad_norm": 1.7152010217915816,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.5929,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.4864864864864865,
132
+ "grad_norm": 1.4638072286809918,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.592,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.5135135135135135,
139
+ "grad_norm": 1.4052441733855927,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.5881,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.5405405405405406,
146
+ "grad_norm": 1.3473253327847512,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.5863,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5675675675675675,
153
+ "grad_norm": 1.6317553588784817,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.5888,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5945945945945946,
160
+ "grad_norm": 1.4742582489032408,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.5846,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.6216216216216216,
167
+ "grad_norm": 1.3786710575715564,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.5858,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.6486486486486487,
174
+ "grad_norm": 1.3706145590107932,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.5823,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6756756756756757,
181
+ "grad_norm": 2.9406828647293684,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5759,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.7027027027027027,
188
+ "grad_norm": 1.874695043037241,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.5788,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.7297297297297297,
195
+ "grad_norm": 1.6258297787646745,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5807,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.7567567567567568,
202
+ "grad_norm": 1.3370174946548334,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.5753,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7837837837837838,
209
+ "grad_norm": 1.4879061874672985,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.5751,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.8108108108108109,
216
+ "grad_norm": 1.4797771423522537,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5752,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.8378378378378378,
223
+ "grad_norm": 1.349831755648445,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5731,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.8648648648648649,
230
+ "grad_norm": 1.487488462220457,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5719,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8918918918918919,
237
+ "grad_norm": 1.3196199401162108,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5736,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.918918918918919,
244
+ "grad_norm": 1.2813534915454334,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.581,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.9459459459459459,
251
+ "grad_norm": 1.2493996252660173,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5726,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.972972972972973,
258
+ "grad_norm": 1.3169592287464336,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5717,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.0,
265
+ "grad_norm": 1.6256059905335656,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.5735,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.0,
272
+ "eval_loss": 0.28570452332496643,
273
+ "eval_runtime": 96.5261,
274
+ "eval_samples_per_second": 103.236,
275
+ "eval_steps_per_second": 0.404,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 1.027027027027027,
280
+ "grad_norm": 1.7234414441453492,
281
+ "learning_rate": 5e-06,
282
+ "loss": 0.4828,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.054054054054054,
287
+ "grad_norm": 1.542357388344519,
288
+ "learning_rate": 5e-06,
289
+ "loss": 0.4753,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.0810810810810811,
294
+ "grad_norm": 1.4861195617131548,
295
+ "learning_rate": 5e-06,
296
+ "loss": 0.4799,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.1081081081081081,
301
+ "grad_norm": 1.4134885594853346,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.4728,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.135135135135135,
308
+ "grad_norm": 1.8446040240127095,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.4775,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.1621621621621623,
315
+ "grad_norm": 1.8467841301436136,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.484,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.1891891891891893,
322
+ "grad_norm": 2.42665171779706,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.4749,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.2162162162162162,
329
+ "grad_norm": 1.8049502845392928,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.48,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.2432432432432432,
336
+ "grad_norm": 2.0761863031810375,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.4809,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.2702702702702702,
343
+ "grad_norm": 2.0655543660952476,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.481,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.2972972972972974,
350
+ "grad_norm": 1.6759566925998968,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.4764,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.3243243243243243,
357
+ "grad_norm": 1.396353193003371,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.4761,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.3513513513513513,
364
+ "grad_norm": 1.397142398326265,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.482,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.3783783783783785,
371
+ "grad_norm": 1.5036245472795235,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.4817,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.4054054054054055,
378
+ "grad_norm": 1.5607076153200972,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.4861,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.4324324324324325,
385
+ "grad_norm": 1.7026202176545633,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.4801,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.4594594594594594,
392
+ "grad_norm": 1.694705438287125,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.4834,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.4864864864864864,
399
+ "grad_norm": 1.8913973826320096,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.4815,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.5135135135135136,
406
+ "grad_norm": 1.3860850720137203,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.4853,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.5405405405405406,
413
+ "grad_norm": 1.6545389203896799,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.4775,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.5675675675675675,
420
+ "grad_norm": 1.6330744148753316,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.484,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.5945945945945947,
427
+ "grad_norm": 1.749422453289866,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.4842,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.6216216216216215,
434
+ "grad_norm": 1.4714668834139133,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.4815,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.6486486486486487,
441
+ "grad_norm": 1.7987332350786904,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.4803,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.6756756756756757,
448
+ "grad_norm": 1.4458046351621974,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.4867,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.7027027027027026,
455
+ "grad_norm": 1.5250625163041274,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.4898,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.7297297297297298,
462
+ "grad_norm": 1.682061525596803,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.4903,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.7567567567567568,
469
+ "grad_norm": 1.725604337448065,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.4832,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.7837837837837838,
476
+ "grad_norm": 1.451360087475008,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.4825,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.810810810810811,
483
+ "grad_norm": 1.6609825444761357,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.4826,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.8378378378378377,
490
+ "grad_norm": 1.9683314087008068,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.4864,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.864864864864865,
497
+ "grad_norm": 1.6782355490104197,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.485,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.8918918918918919,
504
+ "grad_norm": 1.9940863134776443,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.4884,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.9189189189189189,
511
+ "grad_norm": 1.558943375608973,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.4926,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.945945945945946,
518
+ "grad_norm": 1.2455247588814846,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.4855,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.972972972972973,
525
+ "grad_norm": 1.4348648241338484,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.4845,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 2.0,
532
+ "grad_norm": 1.2414831162937485,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.4841,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 2.0,
539
+ "eval_loss": 0.2870478928089142,
540
+ "eval_runtime": 96.6141,
541
+ "eval_samples_per_second": 103.142,
542
+ "eval_steps_per_second": 0.404,
543
+ "step": 740
544
+ },
545
+ {
546
+ "epoch": 2.027027027027027,
547
+ "grad_norm": 2.327501959947698,
548
+ "learning_rate": 5e-06,
549
+ "loss": 0.3857,
550
+ "step": 750
551
+ },
552
+ {
553
+ "epoch": 2.054054054054054,
554
+ "grad_norm": 1.7853879802404622,
555
+ "learning_rate": 5e-06,
556
+ "loss": 0.3828,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 2.081081081081081,
561
+ "grad_norm": 1.6129872542969907,
562
+ "learning_rate": 5e-06,
563
+ "loss": 0.3811,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 2.108108108108108,
568
+ "grad_norm": 1.6822958801213699,
569
+ "learning_rate": 5e-06,
570
+ "loss": 0.379,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 2.135135135135135,
575
+ "grad_norm": 1.5765921764422512,
576
+ "learning_rate": 5e-06,
577
+ "loss": 0.3804,
578
+ "step": 790
579
+ },
580
+ {
581
+ "epoch": 2.1621621621621623,
582
+ "grad_norm": 1.4901475659604642,
583
+ "learning_rate": 5e-06,
584
+ "loss": 0.3877,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 2.189189189189189,
589
+ "grad_norm": 1.6201893013251834,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.3865,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.2162162162162162,
596
+ "grad_norm": 1.5437549027848532,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.3847,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.2432432432432434,
603
+ "grad_norm": 1.8632952369866993,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.392,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.27027027027027,
610
+ "grad_norm": 1.7661982815559183,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.394,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.2972972972972974,
617
+ "grad_norm": 1.7332904607598014,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.3879,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.3243243243243246,
624
+ "grad_norm": 2.252072555591542,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.3904,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.3513513513513513,
631
+ "grad_norm": 1.6253901818998244,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.397,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.3783783783783785,
638
+ "grad_norm": 2.211017287566556,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.3931,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.4054054054054053,
645
+ "grad_norm": 2.398472470432874,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.3938,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.4324324324324325,
652
+ "grad_norm": 1.5682213825969358,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.3916,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.4594594594594597,
659
+ "grad_norm": 2.1044553282329077,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.3881,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.4864864864864864,
666
+ "grad_norm": 1.7564401087246682,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.391,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.5135135135135136,
673
+ "grad_norm": 1.6382220101052676,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.391,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.5405405405405403,
680
+ "grad_norm": 1.4798359424940517,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.3979,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.5675675675675675,
687
+ "grad_norm": 1.4676265261223125,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.3963,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.5945945945945947,
694
+ "grad_norm": 1.6295486852060186,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.399,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.6216216216216215,
701
+ "grad_norm": 1.6192977742906878,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4023,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.6486486486486487,
708
+ "grad_norm": 1.4805451072812197,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4037,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.6756756756756754,
715
+ "grad_norm": 1.6037076836111144,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4037,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.7027027027027026,
722
+ "grad_norm": 1.5225333016382139,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.3945,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.72972972972973,
729
+ "grad_norm": 1.6212777115845565,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.4051,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.756756756756757,
736
+ "grad_norm": 1.5016437573712755,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.406,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.7837837837837838,
743
+ "grad_norm": 1.5902877384401621,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4029,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.810810810810811,
750
+ "grad_norm": 1.3912722358421692,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4055,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.8378378378378377,
757
+ "grad_norm": 1.5492106652733648,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4046,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.864864864864865,
764
+ "grad_norm": 1.5663312579126816,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.4016,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.891891891891892,
771
+ "grad_norm": 1.40383114587804,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.4079,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.918918918918919,
778
+ "grad_norm": 1.6190187870975588,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4049,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.945945945945946,
785
+ "grad_norm": 1.640627877455668,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4065,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.972972972972973,
792
+ "grad_norm": 1.4897415581729503,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4066,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 3.0,
799
+ "grad_norm": 1.4711135774031523,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.408,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 3.0,
806
+ "eval_loss": 0.3071748912334442,
807
+ "eval_runtime": 96.698,
808
+ "eval_samples_per_second": 103.053,
809
+ "eval_steps_per_second": 0.403,
810
+ "step": 1110
811
+ },
812
+ {
813
+ "epoch": 3.027027027027027,
814
+ "grad_norm": 2.6131814809186613,
815
+ "learning_rate": 5e-06,
816
+ "loss": 0.2928,
817
+ "step": 1120
818
+ },
819
+ {
820
+ "epoch": 3.054054054054054,
821
+ "grad_norm": 2.2485831495990847,
822
+ "learning_rate": 5e-06,
823
+ "loss": 0.2851,
824
+ "step": 1130
825
+ },
826
+ {
827
+ "epoch": 3.081081081081081,
828
+ "grad_norm": 1.928587031956844,
829
+ "learning_rate": 5e-06,
830
+ "loss": 0.2828,
831
+ "step": 1140
832
+ },
833
+ {
834
+ "epoch": 3.108108108108108,
835
+ "grad_norm": 1.743854950076673,
836
+ "learning_rate": 5e-06,
837
+ "loss": 0.2809,
838
+ "step": 1150
839
+ },
840
+ {
841
+ "epoch": 3.135135135135135,
842
+ "grad_norm": 1.8950206315568565,
843
+ "learning_rate": 5e-06,
844
+ "loss": 0.2814,
845
+ "step": 1160
846
+ },
847
+ {
848
+ "epoch": 3.1621621621621623,
849
+ "grad_norm": 1.9383522044886479,
850
+ "learning_rate": 5e-06,
851
+ "loss": 0.2865,
852
+ "step": 1170
853
+ },
854
+ {
855
+ "epoch": 3.189189189189189,
856
+ "grad_norm": 1.9203849142483256,
857
+ "learning_rate": 5e-06,
858
+ "loss": 0.2851,
859
+ "step": 1180
860
+ },
861
+ {
862
+ "epoch": 3.2162162162162162,
863
+ "grad_norm": 1.9070432540284974,
864
+ "learning_rate": 5e-06,
865
+ "loss": 0.2874,
866
+ "step": 1190
867
+ },
868
+ {
869
+ "epoch": 3.2432432432432434,
870
+ "grad_norm": 1.9642273546651734,
871
+ "learning_rate": 5e-06,
872
+ "loss": 0.2874,
873
+ "step": 1200
874
+ },
875
+ {
876
+ "epoch": 3.27027027027027,
877
+ "grad_norm": 2.371293282521119,
878
+ "learning_rate": 5e-06,
879
+ "loss": 0.2921,
880
+ "step": 1210
881
+ },
882
+ {
883
+ "epoch": 3.2972972972972974,
884
+ "grad_norm": 2.69733157052314,
885
+ "learning_rate": 5e-06,
886
+ "loss": 0.2891,
887
+ "step": 1220
888
+ },
889
+ {
890
+ "epoch": 3.3243243243243246,
891
+ "grad_norm": 2.4462972161470073,
892
+ "learning_rate": 5e-06,
893
+ "loss": 0.287,
894
+ "step": 1230
895
+ },
896
+ {
897
+ "epoch": 3.3513513513513513,
898
+ "grad_norm": 2.09306288923572,
899
+ "learning_rate": 5e-06,
900
+ "loss": 0.2895,
901
+ "step": 1240
902
+ },
903
+ {
904
+ "epoch": 3.3783783783783785,
905
+ "grad_norm": 2.260854207612739,
906
+ "learning_rate": 5e-06,
907
+ "loss": 0.2865,
908
+ "step": 1250
909
+ },
910
+ {
911
+ "epoch": 3.4054054054054053,
912
+ "grad_norm": 2.692817562351766,
913
+ "learning_rate": 5e-06,
914
+ "loss": 0.2899,
915
+ "step": 1260
916
+ },
917
+ {
918
+ "epoch": 3.4324324324324325,
919
+ "grad_norm": 3.125964731548184,
920
+ "learning_rate": 5e-06,
921
+ "loss": 0.2907,
922
+ "step": 1270
923
+ },
924
+ {
925
+ "epoch": 3.4594594594594597,
926
+ "grad_norm": 2.442243961667573,
927
+ "learning_rate": 5e-06,
928
+ "loss": 0.291,
929
+ "step": 1280
930
+ },
931
+ {
932
+ "epoch": 3.4864864864864864,
933
+ "grad_norm": 2.0307566267966344,
934
+ "learning_rate": 5e-06,
935
+ "loss": 0.2934,
936
+ "step": 1290
937
+ },
938
+ {
939
+ "epoch": 3.5135135135135136,
940
+ "grad_norm": 2.098894756875353,
941
+ "learning_rate": 5e-06,
942
+ "loss": 0.2878,
943
+ "step": 1300
944
+ },
945
+ {
946
+ "epoch": 3.5405405405405403,
947
+ "grad_norm": 2.1484942546619985,
948
+ "learning_rate": 5e-06,
949
+ "loss": 0.2918,
950
+ "step": 1310
951
+ },
952
+ {
953
+ "epoch": 3.5675675675675675,
954
+ "grad_norm": 1.8339116955180341,
955
+ "learning_rate": 5e-06,
956
+ "loss": 0.2943,
957
+ "step": 1320
958
+ },
959
+ {
960
+ "epoch": 3.5945945945945947,
961
+ "grad_norm": 1.9887758275282583,
962
+ "learning_rate": 5e-06,
963
+ "loss": 0.2925,
964
+ "step": 1330
965
+ },
966
+ {
967
+ "epoch": 3.6216216216216215,
968
+ "grad_norm": 1.9845485688385418,
969
+ "learning_rate": 5e-06,
970
+ "loss": 0.2955,
971
+ "step": 1340
972
+ },
973
+ {
974
+ "epoch": 3.6486486486486487,
975
+ "grad_norm": 1.969672930852857,
976
+ "learning_rate": 5e-06,
977
+ "loss": 0.2988,
978
+ "step": 1350
979
+ },
980
+ {
981
+ "epoch": 3.6756756756756754,
982
+ "grad_norm": 1.7604665753983029,
983
+ "learning_rate": 5e-06,
984
+ "loss": 0.2954,
985
+ "step": 1360
986
+ },
987
+ {
988
+ "epoch": 3.7027027027027026,
989
+ "grad_norm": 1.8509109071012708,
990
+ "learning_rate": 5e-06,
991
+ "loss": 0.2962,
992
+ "step": 1370
993
+ },
994
+ {
995
+ "epoch": 3.72972972972973,
996
+ "grad_norm": 1.6839851527778709,
997
+ "learning_rate": 5e-06,
998
+ "loss": 0.2991,
999
+ "step": 1380
1000
+ },
1001
+ {
1002
+ "epoch": 3.756756756756757,
1003
+ "grad_norm": 1.9195065233001387,
1004
+ "learning_rate": 5e-06,
1005
+ "loss": 0.2997,
1006
+ "step": 1390
1007
+ },
1008
+ {
1009
+ "epoch": 3.7837837837837838,
1010
+ "grad_norm": 2.0352568345546262,
1011
+ "learning_rate": 5e-06,
1012
+ "loss": 0.3012,
1013
+ "step": 1400
1014
+ },
1015
+ {
1016
+ "epoch": 3.810810810810811,
1017
+ "grad_norm": 1.8158735733043123,
1018
+ "learning_rate": 5e-06,
1019
+ "loss": 0.3005,
1020
+ "step": 1410
1021
+ },
1022
+ {
1023
+ "epoch": 3.8378378378378377,
1024
+ "grad_norm": 1.9516953165324797,
1025
+ "learning_rate": 5e-06,
1026
+ "loss": 0.3022,
1027
+ "step": 1420
1028
+ },
1029
+ {
1030
+ "epoch": 3.864864864864865,
1031
+ "grad_norm": 1.863346672591352,
1032
+ "learning_rate": 5e-06,
1033
+ "loss": 0.3018,
1034
+ "step": 1430
1035
+ },
1036
+ {
1037
+ "epoch": 3.891891891891892,
1038
+ "grad_norm": 1.8751695571197564,
1039
+ "learning_rate": 5e-06,
1040
+ "loss": 0.3019,
1041
+ "step": 1440
1042
+ },
1043
+ {
1044
+ "epoch": 3.918918918918919,
1045
+ "grad_norm": 1.9801631273484928,
1046
+ "learning_rate": 5e-06,
1047
+ "loss": 0.307,
1048
+ "step": 1450
1049
+ },
1050
+ {
1051
+ "epoch": 3.945945945945946,
1052
+ "grad_norm": 2.003514563060868,
1053
+ "learning_rate": 5e-06,
1054
+ "loss": 0.3058,
1055
+ "step": 1460
1056
+ },
1057
+ {
1058
+ "epoch": 3.972972972972973,
1059
+ "grad_norm": 2.0763911612682837,
1060
+ "learning_rate": 5e-06,
1061
+ "loss": 0.3032,
1062
+ "step": 1470
1063
+ },
1064
+ {
1065
+ "epoch": 4.0,
1066
+ "grad_norm": 2.336230273567409,
1067
+ "learning_rate": 5e-06,
1068
+ "loss": 0.3066,
1069
+ "step": 1480
1070
+ },
1071
+ {
1072
+ "epoch": 4.0,
1073
+ "eval_loss": 0.3498997688293457,
1074
+ "eval_runtime": 96.4986,
1075
+ "eval_samples_per_second": 103.266,
1076
+ "eval_steps_per_second": 0.404,
1077
+ "step": 1480
1078
+ },
1079
+ {
1080
+ "epoch": 4.027027027027027,
1081
+ "grad_norm": 3.0158703873058617,
1082
+ "learning_rate": 5e-06,
1083
+ "loss": 0.1986,
1084
+ "step": 1490
1085
+ },
1086
+ {
1087
+ "epoch": 4.054054054054054,
1088
+ "grad_norm": 2.369544566010332,
1089
+ "learning_rate": 5e-06,
1090
+ "loss": 0.1914,
1091
+ "step": 1500
1092
+ },
1093
+ {
1094
+ "epoch": 4.081081081081081,
1095
+ "grad_norm": 2.183015942133232,
1096
+ "learning_rate": 5e-06,
1097
+ "loss": 0.1906,
1098
+ "step": 1510
1099
+ },
1100
+ {
1101
+ "epoch": 4.108108108108108,
1102
+ "grad_norm": 2.158995110143315,
1103
+ "learning_rate": 5e-06,
1104
+ "loss": 0.1894,
1105
+ "step": 1520
1106
+ },
1107
+ {
1108
+ "epoch": 4.135135135135135,
1109
+ "grad_norm": 2.15897438632279,
1110
+ "learning_rate": 5e-06,
1111
+ "loss": 0.1923,
1112
+ "step": 1530
1113
+ },
1114
+ {
1115
+ "epoch": 4.162162162162162,
1116
+ "grad_norm": 2.3012908533453316,
1117
+ "learning_rate": 5e-06,
1118
+ "loss": 0.1918,
1119
+ "step": 1540
1120
+ },
1121
+ {
1122
+ "epoch": 4.1891891891891895,
1123
+ "grad_norm": 2.3397981392039133,
1124
+ "learning_rate": 5e-06,
1125
+ "loss": 0.1955,
1126
+ "step": 1550
1127
+ },
1128
+ {
1129
+ "epoch": 4.216216216216216,
1130
+ "grad_norm": 2.3318950522877544,
1131
+ "learning_rate": 5e-06,
1132
+ "loss": 0.1967,
1133
+ "step": 1560
1134
+ },
1135
+ {
1136
+ "epoch": 4.243243243243243,
1137
+ "grad_norm": 2.2558016269733017,
1138
+ "learning_rate": 5e-06,
1139
+ "loss": 0.1973,
1140
+ "step": 1570
1141
+ },
1142
+ {
1143
+ "epoch": 4.27027027027027,
1144
+ "grad_norm": 2.0308981605993557,
1145
+ "learning_rate": 5e-06,
1146
+ "loss": 0.2001,
1147
+ "step": 1580
1148
+ },
1149
+ {
1150
+ "epoch": 4.297297297297297,
1151
+ "grad_norm": 2.2105682816818604,
1152
+ "learning_rate": 5e-06,
1153
+ "loss": 0.1999,
1154
+ "step": 1590
1155
+ },
1156
+ {
1157
+ "epoch": 4.324324324324325,
1158
+ "grad_norm": 2.0611643054991777,
1159
+ "learning_rate": 5e-06,
1160
+ "loss": 0.1977,
1161
+ "step": 1600
1162
+ },
1163
+ {
1164
+ "epoch": 4.351351351351352,
1165
+ "grad_norm": 2.244438029380733,
1166
+ "learning_rate": 5e-06,
1167
+ "loss": 0.1988,
1168
+ "step": 1610
1169
+ },
1170
+ {
1171
+ "epoch": 4.378378378378378,
1172
+ "grad_norm": 2.1171487595846683,
1173
+ "learning_rate": 5e-06,
1174
+ "loss": 0.2016,
1175
+ "step": 1620
1176
+ },
1177
+ {
1178
+ "epoch": 4.405405405405405,
1179
+ "grad_norm": 2.129014172266156,
1180
+ "learning_rate": 5e-06,
1181
+ "loss": 0.2045,
1182
+ "step": 1630
1183
+ },
1184
+ {
1185
+ "epoch": 4.4324324324324325,
1186
+ "grad_norm": 2.0849845815435284,
1187
+ "learning_rate": 5e-06,
1188
+ "loss": 0.2049,
1189
+ "step": 1640
1190
+ },
1191
+ {
1192
+ "epoch": 4.45945945945946,
1193
+ "grad_norm": 2.107058909660221,
1194
+ "learning_rate": 5e-06,
1195
+ "loss": 0.2058,
1196
+ "step": 1650
1197
+ },
1198
+ {
1199
+ "epoch": 4.486486486486487,
1200
+ "grad_norm": 2.4419601951125274,
1201
+ "learning_rate": 5e-06,
1202
+ "loss": 0.2063,
1203
+ "step": 1660
1204
+ },
1205
+ {
1206
+ "epoch": 4.513513513513513,
1207
+ "grad_norm": 2.242862620136703,
1208
+ "learning_rate": 5e-06,
1209
+ "loss": 0.2079,
1210
+ "step": 1670
1211
+ },
1212
+ {
1213
+ "epoch": 4.54054054054054,
1214
+ "grad_norm": 2.1964969220779498,
1215
+ "learning_rate": 5e-06,
1216
+ "loss": 0.2063,
1217
+ "step": 1680
1218
+ },
1219
+ {
1220
+ "epoch": 4.5675675675675675,
1221
+ "grad_norm": 2.144927893368169,
1222
+ "learning_rate": 5e-06,
1223
+ "loss": 0.2073,
1224
+ "step": 1690
1225
+ },
1226
+ {
1227
+ "epoch": 4.594594594594595,
1228
+ "grad_norm": 2.4089173007864986,
1229
+ "learning_rate": 5e-06,
1230
+ "loss": 0.2064,
1231
+ "step": 1700
1232
+ },
1233
+ {
1234
+ "epoch": 4.621621621621622,
1235
+ "grad_norm": 2.236620269865094,
1236
+ "learning_rate": 5e-06,
1237
+ "loss": 0.208,
1238
+ "step": 1710
1239
+ },
1240
+ {
1241
+ "epoch": 4.648648648648649,
1242
+ "grad_norm": 2.2273265321363964,
1243
+ "learning_rate": 5e-06,
1244
+ "loss": 0.2091,
1245
+ "step": 1720
1246
+ },
1247
+ {
1248
+ "epoch": 4.675675675675675,
1249
+ "grad_norm": 2.043537478772687,
1250
+ "learning_rate": 5e-06,
1251
+ "loss": 0.2083,
1252
+ "step": 1730
1253
+ },
1254
+ {
1255
+ "epoch": 4.702702702702703,
1256
+ "grad_norm": 2.2566729233438516,
1257
+ "learning_rate": 5e-06,
1258
+ "loss": 0.2099,
1259
+ "step": 1740
1260
+ },
1261
+ {
1262
+ "epoch": 4.72972972972973,
1263
+ "grad_norm": 2.3414348548757853,
1264
+ "learning_rate": 5e-06,
1265
+ "loss": 0.2107,
1266
+ "step": 1750
1267
+ },
1268
+ {
1269
+ "epoch": 4.756756756756757,
1270
+ "grad_norm": 2.231457361542356,
1271
+ "learning_rate": 5e-06,
1272
+ "loss": 0.2096,
1273
+ "step": 1760
1274
+ },
1275
+ {
1276
+ "epoch": 4.783783783783784,
1277
+ "grad_norm": 2.182368419604419,
1278
+ "learning_rate": 5e-06,
1279
+ "loss": 0.2111,
1280
+ "step": 1770
1281
+ },
1282
+ {
1283
+ "epoch": 4.8108108108108105,
1284
+ "grad_norm": 2.3022522015009526,
1285
+ "learning_rate": 5e-06,
1286
+ "loss": 0.2097,
1287
+ "step": 1780
1288
+ },
1289
+ {
1290
+ "epoch": 4.837837837837838,
1291
+ "grad_norm": 2.154576582443656,
1292
+ "learning_rate": 5e-06,
1293
+ "loss": 0.2132,
1294
+ "step": 1790
1295
+ },
1296
+ {
1297
+ "epoch": 4.864864864864865,
1298
+ "grad_norm": 2.185745168624507,
1299
+ "learning_rate": 5e-06,
1300
+ "loss": 0.212,
1301
+ "step": 1800
1302
+ },
1303
+ {
1304
+ "epoch": 4.891891891891892,
1305
+ "grad_norm": 2.1651574566981395,
1306
+ "learning_rate": 5e-06,
1307
+ "loss": 0.2143,
1308
+ "step": 1810
1309
+ },
1310
+ {
1311
+ "epoch": 4.918918918918919,
1312
+ "grad_norm": 2.2200846915512127,
1313
+ "learning_rate": 5e-06,
1314
+ "loss": 0.2137,
1315
+ "step": 1820
1316
+ },
1317
+ {
1318
+ "epoch": 4.945945945945946,
1319
+ "grad_norm": 2.344248992891754,
1320
+ "learning_rate": 5e-06,
1321
+ "loss": 0.2162,
1322
+ "step": 1830
1323
+ },
1324
+ {
1325
+ "epoch": 4.972972972972973,
1326
+ "grad_norm": 2.150842096847866,
1327
+ "learning_rate": 5e-06,
1328
+ "loss": 0.215,
1329
+ "step": 1840
1330
+ },
1331
+ {
1332
+ "epoch": 5.0,
1333
+ "grad_norm": 2.3030280373590335,
1334
+ "learning_rate": 5e-06,
1335
+ "loss": 0.2144,
1336
+ "step": 1850
1337
+ },
1338
+ {
1339
+ "epoch": 5.0,
1340
+ "eval_loss": 0.4100167155265808,
1341
+ "eval_runtime": 95.571,
1342
+ "eval_samples_per_second": 104.268,
1343
+ "eval_steps_per_second": 0.408,
1344
+ "step": 1850
1345
+ },
1346
+ {
1347
+ "epoch": 5.0,
1348
+ "step": 1850,
1349
+ "total_flos": 3097981385441280.0,
1350
+ "train_loss": 0.3954814358659693,
1351
+ "train_runtime": 27395.2752,
1352
+ "train_samples_per_second": 34.555,
1353
+ "train_steps_per_second": 0.068
1354
+ }
1355
+ ],
1356
+ "logging_steps": 10,
1357
+ "max_steps": 1850,
1358
+ "num_input_tokens_seen": 0,
1359
+ "num_train_epochs": 5,
1360
+ "save_steps": 500,
1361
+ "stateful_callbacks": {
1362
+ "TrainerControl": {
1363
+ "args": {
1364
+ "should_epoch_stop": false,
1365
+ "should_evaluate": false,
1366
+ "should_log": false,
1367
+ "should_save": true,
1368
+ "should_training_stop": true
1369
+ },
1370
+ "attributes": {}
1371
+ }
1372
+ },
1373
+ "total_flos": 3097981385441280.0,
1374
+ "train_batch_size": 8,
1375
+ "trial_name": null,
1376
+ "trial_params": null
1377
+ }
training_eval_loss.png ADDED
training_loss.png ADDED